Resilient H<inf>∞</inf> filtering for uncertain linear systems: A descriptor system approach

Abstract

This paper deals with the resilient H ∞ filter design problem for a class of linear discrete-time systems with linear fractional parametric uncertainties. Considering a filter with uncertainties and inaccuracies existing in the implementation process, a augmented filtering error system can be obtained by the descriptor systems approach, which can avoid the coupling terms between system and filter matrices. Concentration of the paper lies in deriving the design conditions of the resilient filter such that the asymptotic stability and the H ∞ performance of the augmented filtering error system can be guaranteed. A common Lyapunov function approach and a parameter-dependent Lyapunov function approach are utilized to solve this problem respectively, and the design conditions are given in term of LMIs. Finally, the simulation results show the validity of the proposed methods.